JMIR Publications

ABSTRACT

Background:

In the U.S., racial disparities in birth outcomes persist and have been widening. Interpersonal and structural racism are leading explanations for the continuing racial disparities in birth outcomes but research to confirm the role of racism and evaluate trends in the impact of racism on health outcomes has been hampered by the challenge of measuring racism. Most of research on discrimination relies on self-reported experiences of discrimination, and there have been few studies examining racial attitudes and bias at the U.S. national level.

Objective:

Investigate associations between state-level Twitter-derived sentiment towards racial/ethnic minorities and birth outcomes

Methods:

We utilized Twitter's Streaming Application Programming Interface (API) to collect 26,027,740 tweets from June 2015 to December 2017 containing at least one race-related term. Sentiment analysis was performed using Support Vector Machines (SVM), a supervised machine learning model. We constructed overall indicators of sentiment towards minorities and sentiment towards race-specific groups. For each year, state-level Twitter derived sentiment towards minorities was merged with births data for that year. The study participants are mothers of singleton births with no congenital abnormalities in 2015-2017 with available data on gestational age (N= 8,369,697) and birth weight (N= 8,367,143). The main outcomes are low birth weight (birthweight ≤ 2499 grams) and preterm birth (gestational age < 37 weeks). We estimated prevalence ratios controlling for individual-level maternal characteristics and state-level demographics using log binomial regression models.

Results:

Accuracy for identifying negative sentiment comparing the machine learning model to manually labeled tweets was 91%. Mothers living in states in the highest tertile of negative sentiment towards racial/ethnic minorities had greater prevalence of low birth weight (+8%, 95% CI: 2%, 13%) and preterm birth (+10%, 95% CI: 0%, 21%) compared to mothers living in the lowest tertile. Sentiment towards minorities, but not sentiment towards non-Hispanic Whites, was associated with adverse birth outcomes among non-Hispanic Whites as well as adverse birth outcomes among racial/ethnic minorities as a group. In stratified subgroup analyses, negative sentiments towards specific racial/ethnic minority groups, Blacks and Asians, predicted adverse outcomes for those groups.

Conclusions:

More negative social context related to race predicted worse birth outcomes for racial/ethnic minorities as well as non-Hispanic Whites.